NPredModels

class jolideco.models.NPredModels(background, calibration=None, *args, **kwargs)

Bases: ModuleDict

Flux components

Parameters:

• background (~torch.Tensor) – Background tensor

• calibration (NPredCalibration) – Calibration model.

Initialize internal Module state, shared by both nn.Module and ScriptModule.

Methods Summary

evaluate(fluxes)	Evaluate npred model
evaluate_per_component(fluxes)	Evaluate npred model per component
from_dataset_numpy(dataset, components[, ...])	Create multiple npred models.

Methods Documentation

evaluate(fluxes)

Evaluate npred model

Parameters:

fluxes (tuple of ~torch.tensor) – Flux components

Returns:

npred_total – Predicted counts tensor

Return type:

~torch.tensor

evaluate_per_component(fluxes)

Evaluate npred model per component

Parameters:

fluxes (tuple of ~torch.tensor) – Flux components

Returns:

npreds – Predicted counts tensor per component

Return type:

dict ~torch.tensor

classmethod from_dataset_numpy(dataset, components, calibration=None)

Create multiple npred models.

Parameters:

• dataset (dict of ~numpy.ndarray) – Dataset

• components (FluxComponents) – Flux components

Returns:

npred_models – NPredModels

Return type:

NPredModel